**** BEGIN LOGGING AT Sun Jun 19 23:59:56 2005 Jun 20 00:22:04 morning Jun 20 00:22:32 i try to update my local mirror and get that error "INFO-Cannot get ChangeSet for revision ..................." Jun 20 00:41:40 good morning all Jun 20 00:43:42 morning Jun 20 00:46:00 hey mr_claus Jun 20 01:04:36 hey bluelightning Jun 20 01:04:40 hi all Jun 20 01:04:46 hi koen Jun 20 01:05:03 hi Jun 20 01:05:56 hi mr_claus Jun 20 01:08:00 hey koen, bluelightning, mr_claus Jun 20 01:08:06 hey mithro Jun 20 01:08:16 hi mithro Jun 20 01:08:17 koen: where you located? Jun 20 01:09:03 mithro: a map says it all: http://gpe.handhelds.org/images/people_eu.png :) Jun 20 01:10:31 we need one of those for all handhelds people Jun 20 01:10:40 the netherlands, about 5 km from germany Jun 20 01:10:52 koen: anything intresting out your way Jun 20 01:11:35 * mithro is still trying to figure out where to go on his way from Amsterdam to Stockholm Jun 20 01:11:35 apart from the Grolsch brewery not much Jun 20 01:12:54 so I've decided to try and track down some people I know on IRC :P maybe buy them a beer or two Jun 20 01:13:43 already met pb at linux.conf.au Jun 20 01:22:15 cool Jun 20 01:23:53 morning alll Jun 20 01:24:00 hey RP Jun 20 01:25:31 A map of the oe/handheld people would be interesting Jun 20 01:26:35 sory if it has asked here but I am stuck with this - http://pastebin.ca/15065 Jun 20 01:27:01 please somebody look at it... Jun 20 01:27:31 RP: your in the UK right? Jun 20 01:31:41 RP: i think I owe you a few beers :P Jun 20 01:33:31 *yawn* Jun 20 01:33:32 good morning Jun 20 01:34:11 hi mickeyl Jun 20 01:34:12 morning Jun 20 01:34:13 hi RP Jun 20 01:34:39 morning Jun 20 01:34:44 morning Jun 20 01:37:16 hey mithro Jun 20 01:39:03 hi Jun 20 01:39:10 hey do13 Jun 20 01:39:21 hi koen Jun 20 01:43:31 i try to update my local mirror and get that error "INFO-Cannot get ChangeSet for revision ..................." **** ENDING LOGGING AT Mon Jun 20 01:47:09 2005 **** BEGIN LOGGING AT Mon Jun 20 01:47:29 2005 Jun 20 01:55:55 good morning pb_ Jun 20 01:56:42 morning mickeyl Jun 20 01:57:56 mickeyl: good morning! Jun 20 01:59:27 hey proti, long time no see. Jun 20 01:59:34 #103 has been fixed Jun 20 01:59:47 Yes, a whole busy weeks (a mess rather). Jun 20 02:00:14 I saw it this morning. Quite puzzling fix imho. Jun 20 02:03:11 hi pb_ Jun 20 02:06:31 mickeyl: Is this the real fix for #103 ? Jun 20 02:07:13 The buildProvider is using its own private pkg_pn without updating the one in BBParseStatus.pkg_pn. Jun 20 02:07:55 real or not, it fixes 103 .) Jun 20 02:08:06 zecke promised to clean up the situation more after 1.3.1 Jun 20 02:08:43 Oh, it does indeed, but its a hack. Jun 20 02:09:18 mickey|meeting: you can try a 'monotone --db=~/mickeyl.db pull ewi546.ewi.utwente.nl openembedded' if you want Jun 20 02:14:43 hi bluelightning Jun 20 02:18:49 observation: montone is faster as bk since the consistency check is much shorter Jun 20 02:22:20 we'll have to see how it scales Jun 20 02:25:42 anybody know how the update command is not working? i want to update the local mirror of OE with the free bitkeeper client Jun 20 02:26:58 mornin' folks Jun 20 02:29:05 hey Twiun Jun 20 02:30:44 morning all Jun 20 02:31:05 hi hrw|work Jun 20 02:31:23 koen: any news on the java front? Jun 20 02:32:03 nope Jun 20 02:33:47 hi Twiun Jun 20 02:57:57 mickey|meeting: we have an empty nslu2-linux monotone server ready to sync with your OE one when it's ready. feel free to send me a public key if you want to be able to push to it. Jun 20 02:59:53 koen: can I send you a key so we can test the syncing? Jun 20 03:00:03 sure Jun 20 03:13:17 i'm already looking at the problem of the update bk:/...... because it's not possible to keep my mirror up-to-date with the server, did i something wrong here? Jun 20 03:19:43 oh, the export is not working too, is there a problem with the server? Jun 20 03:21:35 mickey|meeting: if you send me your monotone key we can see how well it works with 3 repos Jun 20 03:21:46 * koen should use the right window to say stuff Jun 20 03:26:13 rwhitby: monotone is using 4% cpu atm Jun 20 03:26:26 and pushing about 50 KB/s Jun 20 03:27:05 ibot: calc 18 * 4 Jun 20 03:27:19 nm Jun 20 03:27:58 koen: how much data does monotone need to push via net? Jun 20 03:28:18 haven't benchmarked that yet Jun 20 03:28:56 ok Jun 20 03:29:00 koen: got the repo - db is 27MB at my end Jun 20 03:29:06 I think the size of the repo + overhead Jun 20 03:29:30 koen@bitbake:~/OE/monotone/openembedded$ du ~/OE-koen.db -hs Jun 20 03:29:30 26M /home/koen/OE-koen.db Jun 20 03:29:35 * rwhitby now needs to read the monotone manual to work out how to proceed :-) Jun 20 03:29:55 * koen follows rwhitby's example Jun 20 03:30:43 my plan is to copy the files over from nslu2-linux's current BK repo clone of openembedded, then commit that, then push that back Jun 20 03:32:07 using openembedded.rwithby as a branch? Jun 20 03:32:53 actually, I was going to use org.nslu2-linux.rwhitby ... Jun 20 03:33:06 what do you prefer? Jun 20 03:33:35 or maybe just nslu2-linux.rwhitby, or even openembedded.nslu2-linux.rwhitby ... Jun 20 03:33:42 so many choices :-) Jun 20 03:33:52 openembedded.nslu2-linux.rwhitby Jun 20 03:33:58 that's the most descriptive Jun 20 03:34:15 .. Jun 20 03:35:10 and I still haven't messed up my repo :) Jun 20 03:35:46 we (nslu2-linux) will be using @nslu2-linux.org as the id's for writing, so you will always be able to contact a committer directly by looking for .nslu2-linux on freenode.net Jun 20 03:36:17 aha Jun 20 03:37:34 hey lardman Jun 20 03:37:52 hi koen Jun 20 03:38:06 I'll be bk exporting the bk csets to my monotone tree Jun 20 03:38:45 that will mean we'll have a monotone tree ready on july 1st Jun 20 03:39:05 so if we decide on monotone, we'll be prepared Jun 20 03:40:48 koen: nslu2-linux is going to have a central server on mtn.nslu2-linux.org, and our non-core members with write access will write to there. only our core members (the ones with foo.nslu2-linux IRC cloaks) will be requesting write access directly to the oe repo. we will aggregate changes and push as a batch after we review them locally. Jun 20 03:41:21 we currently pull hourly from the oe bk repo into the nslu2-linux openembedded repo Jun 20 03:41:30 sounds pretty sane Jun 20 03:41:49 is it true that you can only serve up one database from monotone serving on the default port? Jun 20 03:41:50 did anyone get mtnweb working yet? Jun 20 03:42:02 rwhitby: it seems so Jun 20 03:42:12 koen: haven't tried, but I'll get our web dude onto it today Jun 20 03:42:28 viewmtn even Jun 20 03:42:41 it's in python, so I'm not going to touch it with a ten foot pole Jun 20 03:43:52 is "mtn" the commonly accepted abbreviation for monotone? Jun 20 03:45:07 * rwhitby is just adding mtn.nslu2-linux.org to the DNS Jun 20 03:46:53 no idea Jun 20 03:48:00 mithro: Yes, I'm in the UK Jun 20 03:48:13 Sorry, was busy with other things :/ Jun 20 03:53:50 Is openzaurus.org down? Jun 20 03:59:11 lardman|work: it's working over here Jun 20 03:59:33 koen: yep, working here too now Jun 20 04:00:06 koen: any idea what the mtn command corresponding to bk sfiles -x -c is ? Jun 20 04:01:05 i get this when using bk : BitKeeper/triggers/post-incoming: Permission denied Jun 20 04:01:26 i can't build the xfce image, on friday i could Jun 20 04:01:55 rwhitby: I was wondering that my self Jun 20 04:02:12 rwhitby: monotone commit should pick up all modified files Jun 20 04:02:44 koen: monotone list unknown Jun 20 04:05:42 ah, very cool Jun 20 04:05:52 intuitive even Jun 20 04:06:03 nicely formatted to pipe to xargs monotone add Jun 20 04:08:55 koen: ready for me to push it back? Jun 20 04:09:19 (on openembedded.nslu2-linux.rwhitby branch) Jun 20 04:09:56 yes Jun 20 04:11:09 done Jun 20 04:11:17 koen@bitbake:~/OE/monotone/openembedded$ monotone list branches Jun 20 04:11:17 openembedded.koen Jun 20 04:11:17 openembedded.nslu2-linux.rwhitby Jun 20 04:11:53 monotone: [bytes in: 6512] [bytes out: 239.2k] [certs out: 4] [revs out: 1] Jun 20 04:12:39 if you bk pull from nslu2-linux.bkbits.net/openembedded you should get the same result Jun 20 04:13:29 and there were many changes and many new files, so it's a good test Jun 20 04:15:24 hmmm, how can I see the diff between to branches> Jun 20 04:15:29 ? Jun 20 04:17:04 * koen reads chapter 2.10 of the manual Jun 20 04:19:56 i have bk, and bb downloaded and correctly installed. i get the oe files using bk, i try to bake xfce-image and i get an error, should i paste the log somewhere to get some help? Jun 20 04:20:36 koen: for write access, are you just listing everyone's id as done in section 2.7 ? Jun 20 04:24:34 Noid: use pastebin.ca Jun 20 04:26:05 hrw|work: and what is that? Jun 20 04:26:40 ~pastebin Jun 20 04:26:44 from memory, pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca Jun 20 04:28:31 rwhitby: I think so Jun 20 04:28:55 http://pastebin.ca/15077 Jun 20 04:29:55 Noid: permission denied - hard to read? Jun 20 04:30:15 lol Jun 20 04:30:44 hrw|work: nope, but i don't understand it, /bin/sh works fine, or so it seems Jun 20 04:31:06 Noid: I never had problem with quilt Jun 20 04:31:17 Noid: what BK tree are you using? Jun 20 04:31:47 the one specified in getting started in the wiki Jun 20 04:31:53 update bk://oe-devel.bkbits.net:8080/openembedded openembedded Jun 20 04:31:55 try that Jun 20 04:32:37 luke-jr_: that is not a problem with repo but with his own system Jun 20 04:32:54 hrw|work: always best to use non-stale stuff for testing Jun 20 04:34:43 koen: just had success restricting write access by matching on the identity using string.find(identity, "@nslu2-linux.org", 1, true) ... Jun 20 04:45:13 so you can embed lua scripts in the conf file? Jun 20 04:48:33 that's pretty neat Jun 20 04:49:26 koen: was the test a success? Jun 20 04:50:23 I have your branch Jun 20 04:50:35 been able to merge it? Jun 20 05:00:02 you mean propagate it? Jun 20 05:00:05 haven't tried that Jun 20 05:01:33 yeah, propagate it Jun 20 05:02:38 * koen monotone propagate openembedded.nslu2-linux.rwhitby openembedded.koen Jun 20 05:02:52 monotone: propagating openembedded.nslu2-linux.rwhitby -> openembedded.koen Jun 20 05:02:53 monotone: [source] 91e238aec3d1d1849b1c4aa6ef2b2886eb36e0e2 Jun 20 05:02:53 monotone: [target] dfec7cfa91fe6bc899e8cf8621e42060072d59c1 Jun 20 05:02:53 monotone: no merge necessary; putting 91e238aec3d1d1849b1c4aa6ef2b2886eb36e0e2 in branch 'openembedded.koen' Jun 20 05:03:25 a ' monotone push ewi546.ewi.utwente.nl openembedded' should get that into the db, right? Jun 20 05:03:39 yeah, should do Jun 20 05:03:47 or just monotone push Jun 20 05:03:58 or even monotone sync Jun 20 05:04:07 (it remembers the other args) Jun 20 05:08:50 monotone? Jun 20 05:09:55 koen: you should also be able to "monotone pull mtn.nslu2-linux.org openembedded.nslu2-linux" and get another copy of what you already have :-) Jun 20 05:11:25 hmm, using cvs sources to download the packages will always be trouble because a lot of times the patches dont fit with the source and nobody know if the sources are changed Jun 20 05:11:53 hmm... my monotone just segfaults =p Jun 20 05:12:10 that's what you get for gentoo Jun 20 05:12:16 yea :| Jun 20 05:12:17 fast and optimized segfaults Jun 20 05:12:50 at least I get something =p Jun 20 05:13:11 I probably wouldn't have half of my system if I was still on LFS Jun 20 05:13:18 mr_claus: that's why in nslu2-linux we always CVSDATE in the distro.conf any packages we use that don't already pull from a tag or a date Jun 20 05:13:43 rwhitby: the nicest thing about it is, that it says "put cset foo in branch bla" Jun 20 05:13:56 so you're not sending the whole patch over the wire Jun 20 05:14:27 for non-merging propagate operations Jun 20 05:15:36 yes, that is very nice Jun 20 05:16:10 rwhitby: only set CVSDATE in conf/distro/openslug.conf? Jun 20 05:16:34 i suspect it would only send the merge info + cset sha1 when merging Jun 20 05:16:40 mr_claus: actually we set it in nslu2.conf (the machine.conf) so that it applies for both unslung and openslug Jun 20 05:17:04 and we have fixeddate snapshots for some packages Jun 20 05:17:50 rwhitby: ok, so there is a problem with gnu-config-native, but it is set to CVSDATE_gnu-config-native ?= "20050221" Jun 20 05:18:07 that's too old Jun 20 05:18:17 thje patches in OE require latest cvs Jun 20 05:19:21 ok, i synced the OE 20 minutes ago and that is configured Jun 20 05:19:35 which date should be used here? Jun 20 05:19:37 mr_claus: time to take this back to #nslu2-linux Jun 20 05:19:53 rwhitby: ok :) Jun 20 05:20:54 mr_claus: any newer then 20050515 Jun 20 05:48:37 rwhitby: is monotone documented in the nslu2 wiki? Jun 20 05:52:19 it would seem that the problem is that my /bin/sh is a symlink to bash, and it doesn't work well with quilt Jun 20 06:00:06 mickeyl: what's the status of your monotone tree? Jun 20 06:03:28 hrw|work: yeah, 20050516 is working :) thx Jun 20 06:06:13 to build openslug as a uclibc based system only to change TARGET_OS="linux" to TARGET_OS="linux-uclibc" in openslug.conf? Jun 20 06:17:31 koen: I'm pretty busy atm., can't set it up today Jun 20 06:17:38 ok, np Jun 20 06:20:39 03koen 07 * r1.3577 10openembedded/packages/gpe-todo/ (gpe-todo_0.54.bb gpe-todo-maemo_0.54.bb): Add gpe-todo 0.54 and gpe-todo-maemo 0.54 Jun 20 06:25:08 rwhitby: that commit should be in the openembedded.nslu2-linux.rwhitby too Jun 20 06:26:44 monotone: successful exchange with ewi546.ewi.utwente.nl Jun 20 06:26:44 monotone: [bytes in: 21.8k] [bytes out: 1295] [revs in: 3] [revs out: 0] Jun 20 06:27:14 monotone: branch 'openembedded.nslu2-linux.rwhitby' is currently merged: Jun 20 06:27:15 c9fa36822a7bc438b032772d161e5e08fa67385c koen@handhelds.org 2005-06-20T13:23:45 Jun 20 06:28:28 it's not as informative as bk when pulling (i.e. you can't scan the changes as they go past like we used to with bk) Jun 20 06:29:10 koen: so I'm not sure how to easily see what has changed since the last pull Jun 20 06:29:27 me neither Jun 20 06:29:35 ok, at least we Jun 20 06:29:42 are both in the dark together :-) Jun 20 06:30:08 I'd love a bk changes -R Jun 20 06:32:46 I love "bk -v -r1.3657 " Jun 20 06:33:14 night 'toners Jun 20 06:34:38 'night rwhitby-asleep Jun 20 06:41:48 03koen 07 * r1.3578 10openembedded/packages/maemo/ (4 files in 2 dirs): update gconf-osso, courtesy Florian Boor Jun 20 07:00:45 while compiling gcc-4.0.0 there is an error, it seems there is missing a patch which is available for gcc-3.4.4 Jun 20 07:00:45 unrecognised em Jun 20 07:00:45 ulation mode: armelf_linux Jun 20 07:00:45 Supported emulations: armelfb_linux armelfb Jun 20 07:08:32 yes, it seems that the arm patches are not available for the gcc-4.0.0 yet Jun 20 07:10:26 how i can force gcc-3.4.4 instead of 4.0.0? Jun 20 07:12:26 mr_claus: PREFERRED_VERSION_gcc-cross = "3.4.4" in config may be Jun 20 07:13:10 gints|wrk: and without that parameter it try to use the newest one? Jun 20 07:13:42 I suppose so... Jun 20 07:18:27 rwhitby-away: when you wake up: http://ewi546.ewi.utwente.nl/tmp/viewmtn/ Jun 20 07:21:08 mickeyl: you'll have to love the python driven web frontend for monotone :) http://ewi546.ewi.utwente.nl/tmp/viewmtn/ Jun 20 07:24:30 koen: quite nice Jun 20 07:25:57 koen: for that maemo stuff maybe maemo.bbclass will be best? gpe-todo-maemo_0.54.bb with only "include gpe-todo_${PV}.bb\n inherit maemo" will be much cleaner? Jun 20 07:26:10 http://ewi546.ewi.utwente.nl/tmp/viewmtn/getdiff.py?id1=91e238aec3d1d1849b1c4aa6ef2b2886eb36e0e2&id2=c9fa36822a7bc438b032772d161e5e08fa67385c Jun 20 07:27:17 that doesn't handle DEPENDS and the different configure flags Jun 20 07:28:46 hmm. right Jun 20 07:29:14 03koen 07 * r1.3579 10openembedded/packages/gpe-todo/gpe-todo.inc: Add missing .inc while nobody is looking Jun 20 07:31:08 koen: i heard you take my name in vain =).. i'll add in libmimic+farsight as soon as something depends on it =) Jun 20 07:31:22 koen: at this rate, its likely to be kopete before gaim :/ Jun 20 07:31:32 lazy gaim people Jun 20 07:31:56 i bet they are too busy annoying gentoo users with bug reports :) Jun 20 07:32:20 http://bugs.gentoo.org/show_bug.cgi?id=35890 Jun 20 07:33:31 koen: link to rss on your blog has pyblogsom.cgi twice;( Jun 20 07:34:09 it has? Jun 20 07:34:10 hmmm Jun 20 07:34:25 I get http://dominion.kabel.utwente.nl/koen/blog/pyblosxom.cgi/index.rss2 Jun 20 07:34:38 http://dominion.kabel.utwente.nl/koen/blog/pyblosxom.cgi/Handhelds/pyblosxom.cgi/index.rss2 Jun 20 07:34:40 * koen blames python and/or Safari Jun 20 07:34:49 etc Jun 20 07:35:13 - that makes it twice.. Jun 20 07:35:35 remove pyblosxom.cgi/ from here Jun 20 07:35:59 ok, fixed Jun 20 07:36:12 koen: :/ Jun 20 07:38:43 hrw|work: maemo is an excellent example why we need USE flags Jun 20 07:39:02 hey jamey_ Jun 20 07:39:40 hi koen Jun 20 07:45:25 OT: movial is hiring Jun 20 07:46:53 hi Jun 20 07:48:28 hey CoreDump|home Jun 20 07:48:40 zecke: nokia too Jun 20 07:48:49 zecke: and o-hand Jun 20 07:49:04 and the Albert Heijn around the corner Jun 20 07:50:00 koen: wow I could see vla? Jun 20 07:50:04 and sell it Jun 20 07:51:39 sure Jun 20 07:51:50 and eat it at the company payed supper Jun 20 07:51:57 Nokia is hiring Qt/E devs? ;) Jun 20 07:52:17 yes, to convert them to symbian or gtk Jun 20 07:52:47 good morning mister Crouse Jun 20 07:54:18 hey proti Jun 20 07:54:38 morning zecke Jun 20 07:56:06 koen: I'll apply at Albert Heijn to sell Vla to bum-heads from duitsland Jun 20 07:58:04 koen: do you hang out on the gimp net as well= Jun 20 08:02:24 only freende Jun 20 08:05:03 03koen 07 * r1.3580 10openembedded/packages/libgpepimc/ (libgpepimc_0.5.bb libgpepimc.inc libgpepimc-maemo_0.6.bb): Add libgpepimc and libgpepimc-maemo 0.5 Jun 20 08:15:19 Hello. Jun 20 08:15:26 kergoth: the package ixp425-eth-1.1-r1 fetch a patch called ixp425_eth.c.patch from sourceforge and it doesn't fit, the first patch part seems to be obsolete Jun 20 08:15:33 hi SirFred Jun 20 08:18:00 * koen is starting to like monotone Jun 20 08:18:38 monotone list unknown | xargs monotone add ; monotone commit --message="foo" ; monotone push Jun 20 08:18:59 very good Jun 20 08:19:04 something for the phrasebook Jun 20 08:21:49 ko - time to go home Jun 20 08:22:12 cu Jun 20 08:23:32 mickeyl: if you send your key, i'll add it, so you can play with it Jun 20 08:24:14 hi Jun 20 08:25:21 koen: which type of key? Jun 20 08:25:46 monotone key Jun 20 08:25:54 can anyone tell me why bitbake on gentoo gives me this error: ERROR: Unable to parse conf/bitbake.conf Jun 20 08:25:55 ah of course Jun 20 08:25:57 ? Jun 20 08:25:59 * mickeyl slaps forehead Jun 20 08:27:00 monotone db init --db=~/mickeyl.db ; monotone --db=~/mickeyl.db genkey Jun 20 08:27:59 mickeyl: http://venge.net/monotone/monotone.html#Creating-a-Database and downwards Jun 20 08:29:03 hm.. Jun 20 08:29:11 Delta_: wrong BBPATH ? Jun 20 08:29:44 BBPATH=/opt/oe/build:/opt/oe/openembedded Jun 20 08:30:02 are you in the build directory when calling bitbake? Jun 20 08:30:04 you should be Jun 20 08:30:11 yup, I am in /opt/oe/build Jun 20 08:30:29 but the bitbake.conf is not in /opt/oe/build/conf but in the /opt/oe/openembedded/conf Jun 20 08:30:42 try bitbake -D -D -D Jun 20 08:31:29 ah it says it can't find conf/auto.conf and conf/site.conf Jun 20 08:31:39 that doesn't matter, those are optional Jun 20 08:32:22 this is the output: Jun 20 08:32:28 DEBUG: CONF reading /opt/oe/openembedded/conf/bitbake.conf Jun 20 08:32:28 DEBUG: CONF file 'conf/site.conf' not found Jun 20 08:32:28 DEBUG: CONF file 'conf/auto.conf' not found Jun 20 08:32:28 DEBUG: CONF including /opt/oe/build/conf/local.conf Jun 20 08:32:28 ERROR: Unable to parse conf/bitbake.conf Jun 20 08:32:36 so it seems to find the file.. Jun 20 08:32:55 are OE and bitbake current ? Jun 20 08:32:57 "unable to parse" can also indicate a type in the configfile Jun 20 08:33:17 echo MACHINE-h3900 > conf/auto.conf will trigger that Jun 20 08:33:21 note the hyphen Jun 20 08:34:26 BitBake Build Tool Core version 1.3.0, bitbake version 1.3.1 Jun 20 08:34:42 and oe is the current version, got it with bk.. Jun 20 08:36:17 use pastebin and show us your local.conf Jun 20 08:37:21 what is pastebin? Jun 20 08:37:26 ~pastebin Jun 20 08:37:28 hmm... pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca Jun 20 08:38:49 ah, ok, one moment.. Jun 20 08:39:36 stinking iptables and 64bit Jun 20 08:39:41 * koen reboots build box Jun 20 08:40:41 I put it there: http://pastebin.ca/15089 Jun 20 08:41:13 aha Jun 20 08:42:04 aha? aha is good? :) Jun 20 08:42:22 * koen hums "take on me" Jun 20 08:43:08 MACHINE = "tune-xscale" Jun 20 08:43:10 that's wrong Jun 20 08:43:12 tune-xscale is not a machine Jun 20 08:44:28 it's not? but it's in conf/machine.. Jun 20 08:44:44 yeah, it's probably misplaced Jun 20 08:44:47 actually I want to build for a pxa270 Jun 20 08:44:47 open up a machine conf Jun 20 08:44:50 and look at the metadata Jun 20 08:45:15 choose a machine that's close to your board Jun 20 08:45:22 and write an own one Jun 20 08:45:38 akita, for example Jun 20 08:45:45 or ipaq-pxa270 Jun 20 08:46:09 I am always a bit scared of those pda's :) Jun 20 08:47:09 write your own machine.conf Jun 20 08:47:13 read my presentation Jun 20 08:47:20 and set at least TARGET_ARCH Jun 20 08:47:28 and PREFERRED_PROVIDER_virtual/kernel Jun 20 08:48:06 one moment, I change it.. Jun 20 08:48:32 ~google building embedded distributions with bitbake and openembedded Jun 20 08:48:48 ibot ? Jun 20 08:48:55 http://vanille.de/tools/FOSDEM2005.pdf Jun 20 08:49:03 heh Jun 20 08:49:06 playing substitute ibot? :) Jun 20 08:49:43 always :) Jun 20 08:50:10 ah you wrote some text about it :) Jun 20 08:50:14 why didn't you say.. Jun 20 08:50:50 isn't this presentation listed among the mandatory things to read before asking questions here? :D Jun 20 08:51:31 ppsst :) Jun 20 08:52:00 Koen: "Phill, is that Mickey?", pb_: "he is talking about openzaurus with a German accent..." Jun 20 08:52:49 heheh Jun 20 08:53:06 i don't have a stinking accent ;) Jun 20 08:53:06 well... unfortunately i do Jun 20 08:55:09 german accents while speaking english sound so bad.. Jun 20 08:55:32 slightly less worse as dutch accents Jun 20 08:55:38 <[g2]> I don't know was you are talk about ! Jun 20 08:55:48 bbl Jun 20 08:56:30 i was on enough international conferences to know that it's possible to understand me... something that can't be said about all attendees Jun 20 08:56:31 heh Jun 20 08:56:45 dutch accents aren't bad, I like dutch accents.. Jun 20 08:59:54 mornin Jun 20 08:59:59 yo treke Jun 20 09:00:02 cya Jun 20 09:00:09 hi treke Jun 20 09:01:18 hey ggilbert Jun 20 09:13:08 mickey|BBQ: enjoy the bbq Jun 20 09:13:23 don't let the meat get too black.. Jun 20 09:13:38 hm, is there some trick about letting bitbake know where to find the .bb files? Jun 20 09:14:07 BBFILES Jun 20 09:14:13 when trying to bitbake nano he keeps telling me that nothing provides nano.. Jun 20 09:14:36 I set BBFILES : BBFILES ="/opt/oe/openembedded/packages/*/*.bb" Jun 20 09:16:42 strange.. Jun 20 09:16:57 no symlinks in the path? Jun 20 09:17:18 nope.. Jun 20 09:20:54 03koen 07 * r1.3581 10openembedded/packages/ (2 files in 2 dirs): Jun 20 09:20:54 libgpepimc.inc: rework libgpepimc.inc, courtesy Florian Boor Jun 20 09:20:54 libgpewidget-hildon_0.102.bb: new file, courtesy Florian Boor Jun 20 09:21:07 RP: ping Jun 20 09:21:15 very confusing.. Jun 20 09:23:09 SirFred: pong Jun 20 09:23:52 RP: Hi. :) Jun 20 09:24:43 RP: A question about framebuffers. I've seen that the framebuffer qte driver gets the linestep from a fb_fix_screeninfo. Jun 20 09:25:07 SirFred: ok Jun 20 09:25:23 SirFred: Keep in mind I'm not an expert in these matters! Jun 20 09:25:26 RP: Should I suppose that given that the linestep property is into the fb_fix_screeninfo, it cannot change it's value? Jun 20 09:25:37 RP: You are more expert than me. Jun 20 09:25:53 I won't change unless the screen changes mode Jun 20 09:26:19 RP: The screen is changing its mode. I think that the problem is that fb_fix_screeninfo is not "reloaded" after the change. Jun 20 09:27:13 RP: I'm seeing that FBIOGET_FSCREENINFO and FFBIOGET_VSCREENINFO are used to fill two structures: finfo and vinfo. Jun 20 09:27:30 RP: Then, vinfo is modified to set the new resolution and pixel depth. Jun 20 09:28:08 RP: Then, they use FBIOPUT_VSCREENINFO to push those changes, and then FBIOGET_VSCREENINFO to reload again vinfo. Jun 20 09:28:29 RP: But they're not reloading finfo using FBIOPUT_FSCREENINFO. Perhaps it's a bug. Jun 20 09:28:44 RP: Excuse me, I mean FBIOGET_FSCREENINFO Jun 20 09:29:30 SirFred: The word fix in there suggests that information should be "fixed" and not subject to change. I know for a fact the w100fb does change information in fix and can't help that Jun 20 09:30:03 SirFred: that does sound like a bug in qte. The xserver does re-fetch the "fix" data after changing resolution due to this exact issue. Jun 20 09:30:26 RP: I'm going to make some changes. Thanks! Jun 20 09:31:08 refetching both does sound a lot safer Jun 20 09:31:24 RP: That's what I've made. Jun 20 09:31:57 RP: And now, to upload the 4Mb library to my poor testing SD. I wonder if I should better use a NFS shared filesystem to test. Jun 20 09:35:00 I've been hammering CF cards for a while now and they're surviving... Jun 20 09:36:00 RP: I'm copying this library at least 15 times a day. Jun 20 09:39:16 SirFred: Think how my Husky feels with the probably thousands of kernels its seen ;-) Jun 20 09:59:44 argh.. I should have read the comments better... lines starting with spaces give parse errors... Jun 20 10:05:11 hey Jun 20 10:06:10 hi zecke Jun 20 10:06:35 I wish it would be rainy, and dark... Jun 20 10:06:44 just like in the U.K. Jun 20 10:07:52 unusually, it hasn't rained here very much today Jun 20 10:08:20 we had a small downpour this morning, but it's been bright and sunny since then. Jun 20 10:09:03 screw england Jun 20 10:09:17 Climate is changing. Jun 20 10:09:19 whatever Jun 20 10:09:30 * zecke starts flashing his simpad image Jun 20 10:09:38 mickey|BBQ: GStreamer is nice Jun 20 10:11:56 ah, zecke's account has been cracked Jun 20 10:12:21 hehe Jun 20 10:12:54 besides being implemented in C (and due this being broken) GStreamer is nice Jun 20 10:20:43 zecke: using c and using glib and still nice ? Jun 20 10:24:53 mallum: the usage of GObject suck ass but there are not too many alternatives around Jun 20 10:25:11 I wonder how jack+gstreamer will perform on an ARM Jun 20 10:27:25 How could I use fbset on openzaurus to change the screen mode? Jun 20 10:27:37 It says me 'unknow video mode' Jun 20 10:28:17 pb_: how can I enlarge the receive buffer of bootldr? Jun 20 10:33:48 zecke: I don't think you can. Jun 20 10:34:09 I'm trying with fbset -xres 480 -yres 640, and it just dumps the new mode info, but nothing seems to change. Jun 20 10:34:12 not without recompiling, anyway Jun 20 10:36:10 Oh, I needed the -n option. Jun 20 10:56:22 bbl.. Jun 20 11:09:07 RP: Did you know that there's a fb_var_screeninfo.rotate field ? Jun 20 11:09:51 RP: It's not in the fb_var_screeninfo that is copied into fbset.c of busybox. It seems a recently added field. Jun 20 11:10:11 RP: Perhaps it's the way to go to set/get rotate state on the w100 Jun 20 12:47:07 evening Jun 20 12:51:38 03zecke 07 * r1.3582 10openembedded/packages/tslib/tslib/simpad/ (tslib.sh ts.conf): Jun 20 12:51:38 tslib (SIMpad): Jun 20 12:51:38 -Use a more approriate device for the 2.6er kernel Jun 20 12:54:12 SirFred: I know about it. I'm not sure how its supposed to be used... Jun 20 12:54:49 RP: OK. I didn't know if you knew about it. Jun 20 12:55:18 Anyone know the / size for C750? Jun 20 12:56:21 SirFred: It was what I said about partial support yesterday - I think there are problems with it Jun 20 12:57:51 RP: Didn't you find any example of use Jun 20 12:59:01 SirFred: No Jun 20 12:59:32 RP: I just got the qte driver to rotate correctly the framebuffer. Jun 20 12:59:50 RP: The only problem is tslib, i'm not sure of how to proceed. Jun 20 13:00:23 bbl Jun 20 13:00:32 SirFred: I really don't know what to suggest there... Jun 20 13:00:44 Perhaps tslib needs a rotation module? Jun 20 13:00:56 RP: I'm not sure. Jun 20 13:01:07 it has a swapxy option Jun 20 13:01:15 but you'll need to do it runtime Jun 20 13:01:44 koen|tv: correct. I'm not even sure swap xy would work :/ Jun 20 13:02:49 Does maemo have a web browser? Jun 20 13:04:16 yes, opera based Jun 20 13:04:54 bleh Jun 20 13:05:08 but nokia is sponsering webcore and minimo too Jun 20 13:05:08 any idea if dillo2 or gtk-webcore have tabs? Jun 20 13:05:34 that mean minimo will become usable now? Jun 20 13:05:49 dillo has per definition no tabs Jun 20 13:06:09 any usefull feature is "bloet" or "fluff" to the dillo people Jun 20 13:06:40 heh Jun 20 13:06:48 * koen|tv wants a way to turn off the small screen css in minimo Jun 20 13:07:30 gtk-webcore is getting better everyday Jun 20 13:08:51 * luke-jr_ wants minimo to have tabs and such Jun 20 13:08:54 (real tabs) Jun 20 13:09:16 if only kdelibs were smaller :\ Jun 20 13:09:57 heh Jun 20 13:10:07 try gnome Jun 20 13:10:13 eww Jun 20 13:10:14 that was seperate libs Jun 20 13:10:35 gnome = ick Jun 20 13:10:41 GPE > GNOME Jun 20 13:11:17 just wait till we have gpe.gnome.org :) Jun 20 13:11:33 * luke-jr_ ponders porting GPE to Qt Jun 20 13:12:03 you mean porting to crap? Jun 20 13:12:28 Qt > GTK Jun 20 13:12:32 to one huge library? Jun 20 13:12:39 Qt is not one huge library Jun 20 13:12:50 anymore Jun 20 13:13:26 hey zecke Jun 20 13:13:33 hey zecke Jun 20 13:13:34 pb_: any idea why CTRL+STRG+F1 stopped switching the VTs for me? Jun 20 13:15:24 zecke: sorry, no idea Jun 20 13:15:29 zecke: Is this on the latest xserver-kdrive? Jun 20 13:15:54 RP: no it is x86 Jun 20 13:16:14 my laptop broke on saturday and I'm still setting it up... Jun 20 13:16:44 zecke: You can run xserver-kdrive on x86 afaik ;-) Jun 20 13:16:58 doing it right now Jun 20 13:17:06 ~hail Xglx Jun 20 13:17:06 RP: I even have an ati card in my laptop Jun 20 13:17:09 vesa is your friend Jun 20 13:17:10 zecke: ah, I think there is an option in your x config file Jun 20 13:17:36 maybe you have Option "DontVTSwitch" set. Jun 20 13:18:15 pb_: sadly not Jun 20 13:18:37 oh well Jun 20 13:18:37 koen|tv: The vt handling on the latest xserver seems a bit broken Jun 20 13:19:06 RP: yeah, breaking stuff is all the rage now Jun 20 13:19:29 And nobody seems to care :-( Jun 20 13:19:42 kdrive bugzilla is a farce Jun 20 13:20:34 koen|tv: yup. I've reported the tslib problem but I have no idea how to fix it... Jun 20 13:20:57 I reported the tslib breakage months ago..... Jun 20 13:20:59 pb_: any ideas why the tslib code in xserver-kdrive would have to init before mice? Jun 20 13:21:29 koen|tv: Nobody had hardware to test it with :-( Jun 20 13:21:31 not offhand. I'd need to look at the code. Jun 20 13:21:49 RP: why in hell did they touch it? Jun 20 13:22:21 pb_: I got that far in tracking it down. I don't know enough about tslib to get further with out a lot of learning about xserver internals :-/ Jun 20 13:22:22 "oooh, we can't test it, so we can do what we want" Jun 20 13:22:32 RP: is there a bug report? if so, I'll take a look tomorrow. Jun 20 13:22:43 pb_: Its in bugzilla Jun 20 13:22:48 you might also see if you can sweet-talk Matthew into looking at it. Jun 20 13:22:59 RP: which bugzilla? Jun 20 13:22:59 we tried :( Jun 20 13:23:01 do you have a url? Jun 20 13:23:12 pb_: I also put a patch into oe which "corrects" the problem by reversing the init order Jun 20 13:23:51 heh Jun 20 13:23:59 pb_: looking at it now Jun 20 13:24:06 ah, there you go Jun 20 13:24:26 mallum: very good Jun 20 13:27:59 pb_: https://bugs.freedesktop.org/show_bug.cgi?id=3563 Jun 20 13:29:02 okay, thanks Jun 20 13:29:50 I'm assuming that mallum has it in hand for the moment, but drop me a line if you want me to take a look. Jun 20 13:30:29 pb_: will do, thanks Jun 20 13:30:49 pb_: just trying to get tslib to place nice on a x86 touchscreen Jun 20 13:30:56 pb_: should be much easier to fix then Jun 20 13:30:58 fwiw, I doubt the problem is anything to do with tslib itself: it's more likely to be an issue with the way the mouse and touchscreen modules within tslib compete for input devices. Jun 20 13:31:00 mallum: good plan Jun 20 13:31:47 pb_: keithp changed the ts code so ts's were treated as mouses but he never tested before committing Jun 20 13:32:15 which was reported months ago Jun 20 13:32:27 I guess you could teach xephyr to generate touchscreen "events" on a socket when you clicked in its window. That might be a cute way to debug this sort of thing. Jun 20 13:34:40 There is also a problem with the vt handling on recent servers. That change went in at the same time as the ts stuff... :/ Jun 20 13:35:19 someone should remove the distracting shiny stuff from keithp's desk Jun 20 13:35:59 I need a suspend/resume to get a working terminal back after exiting X :-( Jun 20 13:43:45 For the record, the arm apm.c looks to be preempt unsafe, just to add to the fun... Jun 20 13:44:42 * koen does not do a little dance Jun 20 13:50:40 Help! I need help with English. Jun 20 13:50:54 "There already is an attribute named apply_entry" <--- Is it correct in English? Jun 20 13:51:52 tigrux: "There is already an attribute..." would be more natural, but what you wrote is also correct. Jun 20 13:52:08 It depends on where you want the stress to fall in the sentence. Jun 20 13:52:26 Oki. Jun 20 13:52:41 pb_: I don't want to overlay attributes to my class. Jun 20 13:54:49 ERROR : flash write by 100%... <-- any idea what that means during a flash? Jun 20 13:57:28 Spyro: power down works as well Jun 20 14:07:42 'night Jun 20 14:07:51 'night SirFred|sleep Jun 20 14:08:28 hmm my whole keymap is messed up Jun 20 14:08:33 'night SirFred|sleep Jun 20 14:39:01 * koen celebrates a year of OE involvement Jun 20 14:40:35 I wonder when I come up to that. Or when we'll have had a year of Zaurus kernels :) Jun 20 14:40:42 Zaurus 2.6 kernels that is Jun 20 16:12:28 03mickeyl 07 * r1.3584 10openembedded/packages/orinoco/ (2 files in 2 dirs): apply patch that tries to fix the firmware loading problem w/ spectrum_cs Jun 20 16:12:31 03mickeyl 07 * r1.3577.1.1 10openembedded/packages/opie-pcmciaapplet/opie-pcmciaapplet.inc: fix opie-pcmciaapplet packaging Jun 20 16:35:56 mickey|zzZZzz: (and koen): when you wake up, I need to talk to you about monotone branch naming conventions. Jun 20 16:43:36 does someone have a Zaurus powersupply and a Voltmeter at hand? Jun 20 17:02:04 * chouimat|ibook is away: food Jun 20 17:48:39 mickey|zzZZzz: koen: pb: have sent you an email on the subject Jun 20 18:01:14 nslu2-linux just got the EOL email from bitkeeper. 11 days until your bk executable ceases to run. Jun 20 18:01:42 is there going to be a cut-off point for new changes to the openembedded bk repo ahead of the switch? Jun 20 18:02:28 and if so, we need to coordinate a pull from bk://nslu2-linux.bkbits.net/openembedded into oe-devel.bkbits.net before the export of the oe data from bk happens. Jun 20 18:03:22 or is the switch going to happen after 1 July, letting bitmover dictate when the last change to the oe repo is accepted ;-) Jun 20 19:38:27 * luke-jr_ begins working on KDE bbs Jun 20 20:30:27 night Jun 20 21:17:50 Hi All, when flashiing an SL-6000 using the new OZ-353 updater.sh, must the CF be VFAT? I still can't get the updater.sh to flash the kernel. Everything else is fine. Jun 20 21:18:10 p.s. I asked on #openzaurus and no-one was home... Jun 20 22:01:18 And the answer is ..... Yes the CF card must be FAT16 Formated for the OZ-353 kernel to install via the new updater.sh Jun 20 23:41:58 hi all! Jun 20 23:49:11 I want to get soap service support (both client and server) on my board - anyone tried csoap? or is axis-c the better solution? **** ENDING LOGGING AT Mon Jun 20 23:59:56 2005